I often hear that Wikipedia should not be taken as necessarily correct or authoritative. Can anyone post some links to current articles that are good examples of why Wiki should not be believed uncritically?
How about the pages listed under Category:NPOV disputes?
Sequence of performances at Woodstock. The Incredible String Band did not perform on the first day (they were originally scheduled to do so, but postponed their set until the second day because of rain); Melanie’s “Birthday of the Sun” is not listed in her set even though the song appeared on the Woodstock 2 album; there are several discrepencies in the order of the acts on the second day compared to more authoritative lists.
In a perfect world, every single error that is mentioned as an answer to the OP’s question should become a former error, right here, right now. If you know that there is an error in a Wikipedia article, and you can prove it, don’t just complain about it on some other message board. There’s a nice handy “edit” button at the top of every article. Please use it. 
*** Ponder
See their page about Paul McCartney, which claims in the very first line that he was born in 1842.
Fortunately, that page has been fixed, and now shows his correct birthdate in 1942.
Nothing should be believed uncritically. Question everything.
That’s how I see it, anyway.
A related point that can be mentioned here is the study, published in Nature in 2005, that concluded Wikipedia’s science articles do not have significantly more errors than Encyclopedia Brittanica.
Of course in Britannica you won’t occasionally find an edition stating that the population of African elephants has tripled in the last decade 
It’s not (as far as I know) an erroneous article itself, but there’s some information about other wikipedia articles (such as elephants) that suffer from occasional innaccuracy at Wikipedia in culture - Wikipedia
I know someone who once completely made up an entire article about an obscure colonial conflict between fishermen and bankers (again, totally made-up). Complete with fake citations, and very well-written in general. He figured it would be received skeptically and discovered quickly as a ruse. Much to all of our amazement, that article stayed there for years (it was just recently deleted as a hoax, as a matter of fact). It becomes point #1 in our arguments for why the accuracy of Wikipedia shouldn’t be blindly trusted.
That study was roundly criticized for not weighting errors according to significance. A quick search of The Register’s website will find the relevant dispute reported.
Very Interesting. The article is here.
Quite damning for Nature. Unfortunately the article does not make it clear where The Register got *their *information from.
Even the above, presumably. Where does that leave us?
You should believe “Nothing should be believed uncritically” to the extent that rational analysis shows it to be a trustworthy principle. The idea isn’t “You should consider everything false”; it’s just “Don’t turn off your critical thinking”.
Here’s the Britannica response to the Nature article (PDF). Interesting reading. Does anyone know, did Nature ever respond to these criticisms?
Also, from the Britannica response:
Did Nature ever make the full reports available?
It’s conceivable that some things should be believed uncritically. Obviously.
A professorial friend of mine commented to me that from what he can tell, most academic subjects’ Wikipedia pages are pretty accurate because the grad students in the various disciplines tend to make vanity-style pages about their focused field of study, and also peer review the others.
As for more generic stuff, from what I can tell, it’s reasonably accurate, but what kills me is when some non-native English speaker writes an article and mangles the language in the process, making the article hard to read and confusing. Then, when someone edits the article to make the fucking thing make sense, they edit it right back to the Engrish version.
Every time I edit a Wiki page, even for a typo, some “editor” changes it back.
I’ve been editing for several years. About 95% of my changes stay with no argument at all.
Tried it. It can’t be done. I’ll give you one examples of why, and they are also, IMO, prime examples of what the OP is looking for.
The article on commercial hemp. You can’t trust a damn thing in there. Almost all the references come from pro-hemp websites or 100 year old material.
It’s persistently full of the usual pro-weed nonsense about hemp not needing herbicides, hemp not having any pests and do forth. A few years ago I decided to try to correct the article. I didn’t remove any claim, I simply inserted references from agriculture journals and agriculture departments worldwide that stated worldwide that hemp isn’t pest free, doesn’t fix nitrogen and so forth. The additions I made were deleted and bowdlerised persistently for several months until I gave up. And today the article seems just as crappy and erroneous as it was then based on the first few paragraphs.
Another saga was prompted by a debate here in SDMB 3 or 4 years ago about whether forests produce oxygen. Someone linked to some Wiki article which claimed they did (reference: some kids education site hosted by a paper mill or something). I corrected the error with references to journals (including Nature), US senate reports, university websites and so forth. But because everybody knows that forests produce oxygen every week when I went back the changes had been removed, usually with a comment like "This can’t be true!!!. After a couple of months I gave up.
And these are both examples of issues of fact that are not disputed by the authorities in the field. But it’s impossible to correct the errors.
But if anybody wants to play a game, I’ll provide some highly reputable references for those claims and we can put them in. And then we’ll see how long it takes before they are removed. Based on experience I say they’ll be gone in a week at the outside.
Essentially on anything that is even vaguely controversial/emotional you can’t trust Wikipedia worth a damn.